This project will utilize metabolic engineering to control the supramolecular assembly known as a bacterial biofilm as well as control virulence by evolving signal receptor proteins (e.g., SdiA, Hha, YmgB, and MqsR) and by utilizing cell signals (e.g., autoinducer-2, indole). Microfluidic devices will be used for secondary screening and to build designer, engineered, multi-species biofilms for applications. The paradigm shift is in controlling biofilms to achieve engineering and medical aims (e.g., biocorrosion, biocatalysis, rhizoremediation, food poisoning) whereas previously biofilms have been studied primarily as a means toward eradicating them. In addition, we aim to control biofilm formation and virulence genes via manipulation of signal regulators rather than try to eliminate the bacterium (i.e., control gene expression via cell signaling rather than discover antimicrobials). We have recently discovered that E. coli and pseudomonads respond to signals they do not synthesize (homoserine lactones influence E. coli biofilms while indole influences those of pseudomonads), that competition for signals is intense to the extent that signals are altered (e.g., indole is hydroxylated by bacteria that do not synthesize it and then regulates a different set of genes), that biofilm signals control pathogenicity loci (e.g., indole, uracil), and that biofilms may be dispersed via global regulators. Here, we will use a simple model system (pathogenic and non-pathogenic Escherichia coli along with pseudomonads and sulfur-reducing bacteria) that allows us to investigate biofilm formation and virulence in a realistic environment (i.e., multi-species biofilms). The novelty of the proposed approach arises from (i) protein engineering of regulatory proteins to control biofilms including formation, dispersal, and virulence (this is one of the first studies to evolve regulators rather than enzymes), (ii) investigation of the concentration-dependent interaction of cell signals, many that we have only recently identified, on biofilm formation, (iii) building designer multi-species biofilms, (iv) and utilizing microfluidic devices to carefully control concentrations and gradients of mixtures of the various signals in multi-species biofilms. In this way, this proposal includes complex biological system (pathogenic/non-pathogenic E. coli, E. coli/sulfur-reducing bacteria, E. coli/ pseudomonads), genomics (DNA microarrays), cell signaling, micro-patterning and microfluidics, and protein engineering (DNA shuffling) along with cell screening (FACS) to tune biofilm formation and cell colonization. If biofilms can be controlled, then they may be used for many diverse applications including reducing corrosion ($276 billion/yr problem in the U.S. or 3% GNP), forming hydrogen for fuel cells, rhizoremediation and biocontrol in agriculture, and patterning in microfluidic devices. If virulence genes can also be controlled in biofilms, then novel treatments can be envisioned for the 80% of bacterial infections that occur in biofilms where antibiotics are often ineffective.